Changes for #11

tidymodels · Sep 13, 2020 · e1167b6 · e1167b6
1 parent 6161326
commit e1167b6
Show file tree

Hide file tree

Showing 4 changed files with 15 additions and 2 deletions.
diff --git a/NEWS.md b/NEWS.md
@@ -4,6 +4,8 @@
 
 * Fixed two bugs in the acquisition function calculations.
 
+* Better control the random number streams in parallel for `tune_grid()` and `fit_resamples()` (#11)
+
 # tune 0.1.1
 
 ## Breaking Changes

diff --git a/R/grid_code_paths.R b/R/grid_code_paths.R
@@ -9,6 +9,7 @@ tune_nothing_with_formula <- function(resamples, grid, workflow, metrics, contro
 # ------------------------------------------------------------------------------
 
 iter_rec_and_mod <- function(rs_iter, resamples, grid, workflow, metrics, control) {
+  set.seed(resamples$.seed[[rs_iter]])
   load_pkgs(workflow)
   load_namespace(control$pkgs)
 
@@ -189,6 +190,7 @@ tune_rec_and_mod <- function(resamples, grid, workflow, metrics, control) {
 # ------------------------------------------------------------------------------
 
 iter_rec <- function(rs_iter, resamples, grid, workflow, metrics, control) {
+  set.seed(resamples$.seed[[rs_iter]])
   load_pkgs(workflow)
   load_namespace(control$pkgs)
 
@@ -330,6 +332,7 @@ tune_mod_with_recipe <- function(resamples, grid, workflow, metrics, control) {
 }
 
 iter_mod_with_recipe <- function(rs_iter, resamples, grid, workflow, metrics, control) {
+  set.seed(resamples$.seed[[rs_iter]])
   load_pkgs(workflow)
   load_namespace(control$pkgs)
 
@@ -470,6 +473,7 @@ tune_mod_with_formula <- function(resamples, grid, workflow, metrics, control) {
 }
 
 iter_mod_with_formula <- function(rs_iter, resamples, grid, workflow, metrics, control) {
+  set.seed(resamples$.seed[[rs_iter]])
   load_pkgs(workflow)
   load_namespace(control$pkgs)
 

diff --git a/R/resample.R b/R/resample.R
@@ -139,6 +139,8 @@ resample_workflow <- function(workflow, resamples, metrics, control) {
 
   has_formula <- has_preprocessor_formula(workflow)
 
+  resamples <- dplyr::mutate(resamples, .seed = sample.int(10^5, nrow(resamples)))
+
   # Save rset attributes, then fall back to a bare tibble
   rset_info <- pull_rset_attributes(resamples)
   resamples <- new_bare_tibble(resamples)
@@ -161,7 +163,7 @@ resample_workflow <- function(workflow, resamples, metrics, control) {
   workflow_output <- set_workflow(workflow, control)
 
   new_resample_results(
-    x = resamples,
+    x = resamples %>% dplyr::select(-.seed),
     parameters = parameters(workflow),
     metrics = metrics,
     outcomes = outcomes,
@@ -205,6 +207,7 @@ resample_with_recipe <- function(resamples, workflow, metrics, control) {
 }
 
 iter_resample_with_recipe <- function(rs_iter, resamples, workflow, metrics, control) {
+  set.seed(resamples$.seed[[rs_iter]])
   load_pkgs(workflow)
   load_namespace(control$pkgs)
 
@@ -337,6 +340,7 @@ resample_with_formula <- function(resamples, workflow, metrics, control) {
 }
 
 iter_resample_with_formula <- function(rs_iter, resamples, workflow, metrics, control) {
+  set.seed(resamples$.seed[[rs_iter]])
   load_pkgs(workflow)
   load_namespace(control$pkgs)
 

diff --git a/R/tune_grid.R b/R/tune_grid.R
@@ -320,6 +320,9 @@ tune_grid_workflow <- function(object,
                      data = resamples$splits[[1]]$data,
                      grid_names = names(grid))
   check_workflow(object, pset = pset)
+
+  resamples <- dplyr::mutate(resamples, .seed = sample.int(10^5, nrow(resamples)))
+
   grid <- check_grid(grid, object, pset)
 
   # Save rset attributes, then fall back to a bare tibble
@@ -340,7 +343,7 @@ tune_grid_workflow <- function(object,
   workflow_output <- set_workflow(object, control)
 
   new_tune_results(
-    x = resamples,
+    x = resamples %>% dplyr::select(-.seed),
     parameters = pset,
     metrics = metrics,
     outcomes = outcomes,